Drugs that have abuse liability in humans typically serve as positive reinforcers to maintain and strengthen behavior leading to their administration in animals and provoke relapse to previously extinguished drug-seeking behavior in abstinent animals. Experiments are being conducted to assess neurobiological and behavioral mechanisms underlying drug self-administration and relapse behavior rats and monkeys, and the ability of pharmacological manipulations to modify such behavior. N-(4-hydroxyphenyl)-arachidonamide (AM404) is an anandamide transport inhibitor shown to reduce rewarding and relapse-inducing effects of nicotine in several animal models of tobacco dependence. However, the reinforcing/rewarding effects of AM404 are not clear. We investigated whether AM404 maintains self-administration behavior or reinstates extinguished drug seeking in squirrel monkeys. In monkeys with a history of anandamide or cocaine self-administration, we substituted injections of AM404 (1100 g/kg/injection). Using a 10-response, fixed-ratio schedule, self-administration behavior was maintained by AM404. Dose-response curves had inverted-U shapes, with peak response rates occurring at a dose of 10 g/kg/injection. In anandamide-experienced monkeys, we also demonstrated self-administration of another anandamide transport inhibitor VDM11. In addition to supporting self-administration, priming injections of AM404 (0.03-0.3 mg/kg) reinstated drug-seeking behavior previously reinforced by cannabinoids (THC or anandamide) or cocaine. Both, AM404 self-administration behavior and reinstatement of drug seeking by AM404 were reduced by treatment with the cannabinoid CB1 receptor antagonist/inverse agonist rimonabant (0.3 mg/kg). Moreover, the reinforcing effects of AM404 were potentiated by the treatment with the FAAH inhibitor URB597 (0.3 mg/kg) suggesting the major role of anandamide in these effects. Finally, AM404 (0.3 mg/kg) potentiated the reinforcing effects of anandamide, but not those of cocaine. In non-human primates, AM404 effectively reinforced self-administration behavior and induced reinstatement of drug-seeking behavior in abstinent monkeys. These effects appeared to be mediated by cannabinoid CB1 receptors. Therefore, compounds that promote actions of endocannabinoids throughout the brain by inhibiting their membrane transport may have a potential for abuse. Nicotine, the main psychoactive component of tobacco, and (-)-(9)-tetrahydrocannabinol (THC), the main psychoactive ingredient in cannabis, play major roles in tobacco and marijuana dependence as reinforcers of drug-seeking and drug-taking behavior. Drugs that act as inverse agonists of cannabinoid CB1 receptors in the brain can attenuate the rewarding and abuse-related effects of nicotine and THC, but their clinical use is hindered by potentially serious side effects. The recently developed CB1-receptor neutral antagonists may provide an alternative therapeutic approach to nicotine and cannabinoid dependence. We compared the attenuation of nicotine and THC reinforcement and reinstatement in squirrel monkeys by the CB1-receptor inverse agonist rimonabant and by the recently developed CB1-receptor neutral antagonist AM4113. Both rimonabant and AM4113 reduced two effects of nicotine and THC that play major roles in tobacco and marijuana dependence: (1) maintenance of high rates of drug-taking behavior, and (2) priming- or cue-induced reinstatement of drug-seeking behavior in abstinent subjects (models of relapse). In contrast, neither rimonabant nor AM4113 modified cocaine-reinforced or food-reinforced operant behavior under similar experimental conditions. However, both rimonabant and AM4113 reduced cue-induced reinstatement in monkeys trained to self-administer cocaine, suggesting the involvement of a common cannabinoid-mediated mechanism in the cue-induced reinstatement for different drugs of abuse. These findings point to CB1-receptor neutral antagonists as a new class of medications for treatment of both tobacco dependence and cannabis dependence. The use cannabis is typically initiated during adolescence. The endocannabinoid system has an important role in formation of the nervous system, from very early development through adolescence. Cannabis exposure during this vulnerable period might lead to neurobiological changes that affect adult brain functions and increase the risk of cannabis use disorder. The aim of this study was to investigate whether exposure to (9)-tetrahydrocannabinol (THC) in adolescent rats might enhance reinforcing effects of cannabinoids in adulthood. Male adolescent rats were treated with increasing doses of THC (or its vehicle) twice/day for 11 consecutive days (PND 45-55). When the animals reached adulthood, they were tested by allowing them to intravenously self-administer the cannabinoid CB1-receptor agonist WIN55,212-2. In a separate set of animals given the same THC (or vehicle) treatment regimen, electrophysiological and neurochemical experiments were performed to assess possible modifications of the mesolimbic dopaminergic system, which is critically involved in cannabinoid-induced reward. Behavioral data showed that acquisition of WIN55,212-2 self-administration was enhanced in THC-exposed rats relative to vehicle-exposed controls. Neurophysiological data showed that THC-exposed rats displayed a reduced capacity for WIN55,212-2 to stimulate firing of dopamine neurons in the ventral tegmental area and to increase dopamine levels in the nucleus accumbens shell. These findings-that early, passive exposure to THC can produce lasting alterations of the reward system of the brain and subsequently increase cannabinoid self-administration in adulthood-suggest a mechanism by which adolescent cannabis exposure could increase the risk of subsequent cannabis dependence in humans.